← Papers
Graphs

MP-PDE

Message Passing Neural PDE Solvers

Paper title Message Passing Neural PDE Solvers

Brandstetter et al. · General PDEs · Public: code

Paper arXiv 2202.03376PDF

Every part of the solver is a graph network passing messages between grid points; it predicts several future time steps at once (temporal bundling) and is trained on inputs produced by its own previous prediction (the pushforward trick) to stay stable over long rollouts.

Published
ICLR

Share on LinkedIn, X or Bluesky, or or .

Know related work that belongs here?

Something wrong on this page?

More papers

All papers
  1. GraphsR-BiLoGo — Transfer learning for rib effects in sheet-forming surrogates
  2. GraphsRead-Write-Relax — Why neural PDE surrogates need both global and local processing
  3. GraphsDGN / LDGN — Sampling flow distributions on meshes with diffusion graph networks